US20080228093A1 - Systems and methods for enhancing cardiac signal features used in morphology discrimination - Google Patents

Abstract

Methods and devices used to classify cardiac events based on morphological analysis of sensed signals are described. A signal comprising a cardiac signal component and a noise signal component is sensed. The sensed signal is processed to preferentially alter morphology of the cardiac signal component. The altered morphology of the cardiac signal component enhances detection of one or more features of the cardiac signal component. The features of the cardiac signal component are detected and the cardiac event is classified using the detected features. Processing the sensed signal may involve the use of adaptable signal processing parameters. For example, the signal processing parameters may be selected to accentuate one or more desirable features of the cardiac signal component or to mitigate one or more undesirable features of the cardiac signal component.

Description

FIELD OF THE INVENTION
• The present invention relates generally to medical devices and, more particularly, to cardiac devices and methods used for classification of cardiac events based on morphological analysis of cardiac signals.
BACKGROUND OF THE INVENTION
• When functioning normally, the heart produces rhythmic contractions and is capable of pumping blood throughout the body. However, due to disease or injury, the heart rhythm may become irregular resulting in diminished pumping efficiency. Arrhythmia is a general term used to describe heart rhythm irregularities arising from a variety of physical conditions and disease processes. Cardiac rhythm management systems, such as implantable pacemakers and cardiac defibrillators, have been used as an effective treatment for patients with serious arrhythmias. These systems typically include circuitry to sense electrical signals from the heart and a pulse generator for delivering electrical stimulation pulses to the heart. Leads extending into the patient's heart are connected to electrodes that contact the myocardium for sensing the heart's electrical signals and for delivering stimulation pulses to the heart in accordance with various therapies for treating the arrhythmias.
• Cardiac rhythm management systems may pace the heart by stimulating the heart tissue to produce a contraction of the tissue. Pacemakers deliver a series of low energy pace pulses timed to assist the heart in producing a contractile rhythm that maintains cardiac pumping efficiency. Detecting if a pacing pulse “captures” the heart and produces a contraction allows the cardiac rhythm management system to adjust the energy level of pace pulses to correspond to the optimum energy expenditure that reliably produces capture. A pace pulse must exceed a minimum energy value, or capture threshold, to produce a contraction. It is desirable for a pace pulse to have sufficient energy to stimulate capture of the heart without expending energy significantly in excess of the capture threshold.
• Cardiac rhythm management systems may also detect and terminate abnormal tachyarrhythmias using a variety of tiered therapies. These tiered therapies range from the delivery of anti-tachyarrhythmia pacing (ATP) to high energy shocks to terminate tachyarrhythmia or fibrillation. To effectively deliver these treatments, the CRM device must first identify the type of arrhythmia that is occurring.
• Detecting various cardiac events, such as capture or a tachyarrhythmia event, may be accomplished by analyzing the morphology of a cardiac signal, such as an electrogram (EGM) or electrocardiogram (ECG). The morphological analysis may involve determining the presence and/or location of cardiac signal features in the cardiac signal. This type of analysis is simplified when the cardiac signal features used to detect the cardiac events are relatively consistent and can be discriminated from other signal features and from noise.
• There is a need for methods and systems that enhance detection of cardiac signal features used to classify various cardiac events. The present invention fulfills these and other needs.
SUMMARY OF THE INVENTION
• The present invention is directed to methods and devices used to classify cardiac events based on morphological analysis of sensed signals. One embodiment of the invention involves a method for classifying a cardiac event. The method includes sensing a signal comprising a cardiac signal component and a noise signal component. The sensed signal is processed to preferentially alter morphology of the cardiac signal component. The altered morphology of the cardiac signal component enhances detection of one or more features of the cardiac signal component. The one of more features of the cardiac signal component are detected and the cardiac event is classified using the detected features.
• In certain implementations, processing the sensed signal involves processing the sensed signal using adaptable signal processing parameters. The signal processing parameters may be automatically determined or may be manually determined by a physician. For example, the signal processing parameters may be selected to accentuate one or more desirable features of the cardiac signal component or to mitigate one or more undesirable features of the cardiac signal component.
• According to one aspect, the signal processing parameters may be determined to enhance a correlation coefficient between beats from cardiac episodes and a template.
• Another aspect of the invention involves detecting one or more undesirable features of the cardiac signal component and determining signal processing parameters that mitigate the undesirable features of the cardiac signal component. For example, the undesirable features may be detected using a gradient analysis or using a curvature analysis.
• The undesirable features may be detected and/or the signal processing parameters enhanced based on signals produced using a pacing protocol (e.g., AAI pacing to simulate a tachyarrhythmia episode) and/or may be based on stored cardiac signals.
• According to various applications, the cardiac events classified using the altered cardiac signal component may include cardiac pacing responses and/or tachyarrhythmia episodes.
• Another embodiment of the invention is directed to a medical system used to classify cardiac events. The system includes sensing circuitry configured to sense a signal comprising a cardiac signal component and a noise component. Signal processing circuitry is configured to process the sensed signal to preferentially alter morphology of the cardiac signal component of the sensed signal. The altered cardiac signal component enhances detection of one or more features of the cardiac signal component. A feature detector is configured to detect the one or more features of the cardiac signal component. A cardiac event processor configured to classify a cardiac event using the detected features.
• A filter may be used to alter the cardiac signal component. In one example, the filter includes a high pass corner frequency of about 3 Hz and a low pass corner frequency of about 100 Hz. The system may additionally include a noise filter configured to filter the sensed signal to reduce the noise component of the sensed signal.
• Parameters of the signal processing circuitry may be derived from population data or may be based on individual characteristics of a patient. The parameters of the signal processing circuitry can be automatically adapted by the device or programmed by a physician.
• According to one implementation, the signal processing circuitry comprises a bank of signal processing units. Each of the signal processing units may have associated signal processing parameters and/or may be optimized to recognize or classify a particular type cardiac event.
• The above summary of the present invention is not intended to describe each embodiment or every implementation of the present invention. Advantages and attainments, together with a more complete understanding of the invention, will become apparent and appreciated by referring to the following detailed description and claims taken in conjunction with the accompanying drawings.
BRIEF DESCRIPTION OF THE DRAWINGS
• FIG. 1 is a flow graph of a method for identification of cardiac events using signal processing for enhanced feature detection in accordance with embodiments of the invention;
• FIGS. 2A and 2B are graphs that illustrate modifying the cardiac signal component of a sensed signal to enhance feature detection in accordance with embodiments of the invention;
• FIGS. 3A-3D illustrate the improvement in accurate identification of a captured response to cardiac pacing when the cardiac signal component of the sensed signal is altered for enhanced feature detection in accordance with embodiments of the invention;
• FIG. 4 shows an implantable cardiac rhythm management (CRM) device that may be used to implement signal processing to enhance feature detection in accordance with the present invention;
• FIG. 5A presents a functional block diagram of CRM circuitry configured to implement signal processing to enhance feature detection and cardiac event identification in accordance with embodiments of the invention;
• FIG. 5B is a block diagram illustrating a bank of signal processing units where each signal processing unit is used to enhance cardiac signal features used in identification of a particular type of cardiac event in accordance with embodiments of the invention;
• FIG. 6 is a flow graph that illustrates a method for automatically determining signal processing parameters for feature enhancement in accordance with embodiments of the invention;
• FIG. 7A is a graph of a cardiac signal prior to processing, where the cardiac signal has a morphology exhibiting double positive peaks;
• FIG. 7B is a graph of the gradient of the cardiac signal ofFIG. 7A including zero crossings indicating undesirable multiple peaks;
• FIG. 7C is a graph of the cardiac signal after processing illustrating the reduction of the double peak morphology ofFIG. 7A;
• FIG. 7D is a graph of the gradient of the cardiac signal ofFIG. 7C illustrating a single zero crossing corresponding to the single peak of the cardiac signal;
• FIG. 8 is a graph of a cardiac signal illustrating feature points extracted for cardiac rhythm identification; and
• FIG. 9 illustrates windows drawn around feature point locations1-8. These windows may be used to detect significant changes in curvature and/or the presence of multiple inflection points within the window area in accordance with embodiments of the invention.
• While the invention is amenable to various modifications and alternative forms, specifics thereof have been shown by way of example in the drawings and will be described in detail below. It is to be understood, however, that the intention is not to limit the invention to the particular embodiments described. On the contrary, the invention is intended to cover all modifications, equivalents, and alternatives falling within the scope of the invention as defined by the appended claims.
DETAILED DESCRIPTION OF VARIOUS EMBODIMENTS
• In the following description of the illustrated embodiments, references are made to the accompanying drawings forming a part hereof, and in which are shown by way of illustration, various embodiments by which the invention may be practiced. It is to be understood that other embodiments may be utilized, and structural and functional changes may be made without departing from the scope of the present invention.
• Detection of various cardiac events and/or conditions, including cardiac pacing response events and/or tachyarrhythmia events may be accomplished through morphological analysis of the cardiac signal produced by the heart. The cardiac signal may be acquired via implanted electrodes coupled to a pacemaker, defibrillator, or other type of cardiac rhythm management system. Various types of cardiac events are associated with consistently occurring cardiac signal features that can be used to classify the type of cardiac event.
• Detection of cardiac signal features used to classify cardiac events may be made more difficult due to morphological characteristics of the cardiac signal. For example, undesirable cardiac signal features that occur close in time to a feature used for event identification are problematic. These undesirable features cause variation in discriminating the feature used for event identification. For example, if cardiac event identification is based on the amplitude and timing of a cardiac signal peak, a cardiac signal exhibiting a single sharp peak is desirable. The presence of a double peak or a flattened peak may lead to variation in determining the peak coordinates and corresponding inaccuracy in event identification.
• The flow graph ofFIG. 1 illustrates a method involving cardiac event identification using signal processing for enhanced feature detection in accordance with some embodiments. The method involves sensing110 a signal that includes a cardiac signal component and a noise signal component. The sensed signal may derive from an electrical source, such as cardiac electrical activity, or a non-electrical source, such as a pressure signal, a respiration signal, or heart sound signal. The sensed signal is processed120 to preferentially alter the morphology of the cardiac signal component. Preferential alteration of the cardiac signal component may involve, for example, signal processing to accentuate one or more features of the cardiac signal component and/or to reduce one or more undesirable features of the cardiac signal component. The preferential alteration of the cardiac signal component does not necessarily modify the noise component. In some implementations, the noise component is filtered in a separate signal processing step. After alteration of the cardiac signal component, one or more features are detected130 in the cardiac signal component. A cardiac event is identified140 using the detected features.
• A signal sensed via implantable or patient-external sensors, such as cardiac electrodes, pressure sensors, respiration sensors, accelerometers, or other sensors, generally includes a noise component superimposed on the cardiac signal component. For example, if the sensed signal is related to cardiac activity, the resultant composite signal is thus not necessarily an accurate representation of cardiac activity because the noise signal component is also present in the composite signal. For purposes of this discussion, noise signals are signals from sources other than the cardiac activity of the heart. Noise sources may include environmental noise, such as 60 Hz power line noise, myopotentials from skeletal muscle, motion artifacts, and/or other noise sources. Much effort has previously been directed to removing noise from the cardiac signal to achieve the “pure” cardiac signal component without noise or with a reduced level of noise. Reduction of noise is one technique used to facilitate morphological analysis of sensed signals to classify cardiac events. For example, increasing the signal to noise ratio of the sensed signal by filtering the noise component can enhance feature detection.
• Noise reduction approaches attempt to reveal the cardiac signal from a signal that combines the cardiac signal with signal components from unwanted sources. The approaches of the present invention are directed to preferentially modifying the cardiac signal component of the sensed signal to enhance signal feature detection rather than to achieve the reduction of noise in the sensed signal. Modification of the cardiac signal component as illustrated by various embodiments herein may be accomplished with or without the process of noise reduction. For example, the signal processing techniques of the present invention may be applied to change the shape of the cardiac signal component to enhance detection of signal features present in the cardiac signal component of a sensed signal.
• Signal processing for feature enhancement of the cardiac signal in accordance with embodiments of the invention may be achieved by processing the signal sensed via cardiac electrodes using one or more signal processors. The graphs ofFIGS. 2A and 2B illustrate modification of the cardiac signal component to enhance feature detection.FIG. 2A illustrates acardiac signal210 exhibiting a relatively sharpnegative peak220 and a flattenedpositive peak230.FIG. 2B illustrates the modifiedcardiac signal240 after processing the signal to sharpen thepositive peak250.
• Signals processed to enhance feature detection lead to more accurate identification of cardiac events.FIGS. 3A-3D illustrate the improvement in accurate identification of a captured response to cardiac pacing.FIGS. 3A and 3C illustrate a number ofcardiac signals305,306 sensed following pacing. A captured response is identified by negative and positive signal peaks that occur withindetection windows310,320.Circles311,321 ofFIGS. 3B and 3D indicate the location of the cardiac signal peaks of thevarious signals305,306. As can be seen inFIG. 3A, theunprocessed signals305 initially exhibit positive peaks having a somewhat flattened or double peak morphology. This morphology leads to group of signals havingpositive peaks312 that fall outside thecapture detection window310. Thesignals having peaks312 falling outside thecapture detection window310 would be incorrectly identified as a response other than a captured response.
• After processing, thecardiac signals306 exhibit altered morphology that enhances positive peak detection, as illustrated inFIGS. 3C and 3D. Thesignals306 illustrated inFIG. 3C have been processed, e.g., filtered, to alter the morphology of thesignals306.FIG. 3D shows that the locations of thepositive peaks311 now fall within thecapture detection window310 resulting in accurate capture detection.
• A system capable of processing signals to enhance feature detection may be implemented in a cardiac rhythm management (CRM) device such as an implantable cardiac defibrillator, pacemaker or resynchronization device. Although the present system is described in conjunction with device having a microprocessor-based architecture, it will be understood that the CRM device may be implemented in any logic-based integrated circuit architecture, if desired. Furthermore, methods of the present invention may be implemented in a patient-external device or a system that incorporates both patient-external and implantable components.
• Referring now toFIG. 4 of the drawings, there is shown a CRM device that may be used to implement signal processing to enhance feature detection in accordance with the present invention. The cardiac rhythm management system inFIG. 4 includes aCRM device400 electrically and physically coupled to alead system402. The housing and/or header of theCRM device400 may incorporate one ormore electrodes408,409 used to provide electrical stimulation energy to the heart and to sense cardiac electrical activity. TheCRM device400 may utilize all or a portion of the housing as acan electrode409. Thedevice400 may include anindifferent electrode408 positioned, for example, on the header or the housing of thedevice400. If thedevice400 includes both a can electrode409 and anindifferent electrode408, theelectrodes408,409 typically are electrically isolated from each other.
• Thelead system402 includes implantable electrodes used to detect electric cardiac signals produced by the heart and to provide electrical energy to the heart under certain predetermined conditions to treat cardiac arrhythmias. Thelead system402 may include one or more electrodes used for pacing, sensing, and/or defibrillation. In the embodiment shown inFIG. 4, thelead system402 includes an intracardiac right ventricular (RV)lead system404, an intracardiac right atrial (RA)lead system405, an intracardiac left ventricular (LV)lead system406, and an extracardiac left atrial (LA)lead system410. Thelead system402 ofFIG. 4 illustrates one embodiment that may be used in connection with the signal feature enhancement methodologies described herein. Other leads and/or electrodes may additionally or alternatively be used.
• Thelead system402 may include intracardiac leads404,405,406 inserted into a patient's heart and electrically coupled to the cardiac myocardium. The intracardiac leads404,405,406 include various electrodes positionable within the heart for sensing electrical activity of the heart and for delivering electrical stimulation energy to the heart, for example, pacing pulses and/or defibrillation shocks to treat various arrhythmias of the heart.
• As illustrated inFIG. 4, thelead system402 may include one or more extracardiac leads410 having electrodes, e.g., epicardial electrodes, positioned at locations outside the heart for sensing and pacing one or more heart chambers.
• The rightventricular lead system404 illustrated inFIG. 4 includes an SVC-coil416, an RV-coil414, an RV-ring electrode411, and an RV-tip electrode412. The rightventricular lead system404 extends through the right atrium420 and into the right ventricle419. In particular, the RV-tip electrode412, RV-ring electrode411, and RV-coil electrode414 are positioned at appropriate locations within the right ventricle for sensing and delivering electrical stimulation pulses to the heart. The SVC-coil416 is positioned at an appropriate location within the right atrium chamber of the heart or a major vein leading to the right atrial chamber of the heart.
• In one configuration, the RV-tip electrode412 referenced to thecan electrode409 may be used to implement unipolar pacing and/or sensing in the right ventricle419. Bipolar pacing and/or sensing in the right ventricle may be implemented using the RV-tip412 and RV-ring411 electrodes. In yet another configuration, the RV-ring411 electrode may optionally be omitted, and bipolar pacing and/or sensing may be accomplished using the RV-tip electrode412 and the RV-coil414, for example. The RV-coil414 and the SVC-coil416 are defibrillation electrodes.
• Theleft ventricular lead406 includes an LVdistal electrode413 and an LVproximal electrode417 located at appropriate locations in or about the left ventricle and used for pacing and/or sensing the electrical signals of the left ventricle. Theleft ventricular lead406 may be guided into the right atrium of the heart via the superior vena cava. From the right atrium, theleft ventricular lead406 may be deployed into the coronary sinus ostium, the opening of thecoronary sinus450. Thelead406 may be guided through thecoronary sinus450 to a coronary vein of the left ventricle. This vein is used as an access pathway for leads to reach the surfaces of the left ventricle which are not directly accessible from the right side of the heart. Lead placement for theleft ventricular lead406 may be achieved via subclavian vein access and a preformed guiding catheter for insertion of theLV electrodes413,417 adjacent to the left ventricle.
• Unipolar pacing and/or sensing in the left ventricle may be implemented, for example, using the LV distal electrode referenced to thecan electrode409. The LVdistal electrode413 and the LVproximal electrode417 may be used together as bipolar sense and/or pace electrodes for the left ventricle. Pacing delivered to the heart via theleft ventricular lead406 and theright ventricular lead404 may be used to provide cardiac resynchronization therapy such that the ventricles of the heart are paced substantially simultaneously, or in phased sequence, to provide enhanced cardiac pumping efficiency for patients suffering from chronic heart failure.
• The rightatrial lead405 includes a RA-tip electrode456 and an RA-ring electrode454 positioned at appropriate locations in the right atrium for sensing and pacing the right atrium. In one configuration, the RA-tip456 referenced to thecan electrode409, for example, may be used to provide unipolar pacing and/or sensing in the right atrium. In another configuration, the RA-tip electrode456 and the RA-ring electrode454 may be used to provide bipolar pacing and/or sensing.
• FIG. 4 illustrates one embodiment of a leftatrial lead system410. In this example, the leftatrial lead410 is implemented as an extracardiac lead with LA distal418 and LA proximal415 electrodes positioned at appropriate locations outside the heart for sensing and pacing the left atrium. Unipolar pacing and/or sensing of the left atrium may be accomplished, for example, using the LAdistal electrode418 to thecan409 pacing vector. The LA proximal415 and LA distal418 electrodes may be used together to implement bipolar pacing and/or sensing of the left atrium.
• Circuitry575 used to classify cardiac events and to control the delivery of therapy is enclosed within the housing of thedevice400.FIG. 5A presents a functional block diagram ofCRM circuitry575 in accordance with one embodiment. It is understood by those skilled in the art that there exist many possible configurations in which these functional blocks can be arranged. The exemplary device depicted inFIG. 5A is one possible functional arrangement. Other arrangements are also possible. For example, more, fewer or different functional blocks may be used to describe a device suitable for implementing the methodologies of the present invention.
• TheCRM device circuitry575 is typically powered by an electrochemical battery (not shown). Amemory545 stores data and program commands used to implement cardiac event identification according to the embodiments of the present invention as well as other operations such as therapy delivery. Data and program commands may be transferred between thedevice circuitry575 and a patient-external device555 via telemetry-basedcommunications circuitry550.
• Thecircuitry575 includes atherapy control processor565 capable of controlling the delivery of pacing pulses and/or defibrillation shocks to the right ventricle, left ventricle, right atrium and/or left atrium. The pacingpulse generator530 is configured to generate pacing pulses for treating bradyarrhythmia, for example, or for synchronizing the contractions of contralateral heart chambers using biatrial and/or biventricular pacing. Furthermore, under control of thetherapy control processor565, the cardioversion/defibrillation pulse generator535 may be used to generate high energy shocks to terminate tachyarrhythmia episodes.
• The pacing pulses and/or defibrillation shocks are delivered via multiplecardiac electrodes505 disposed at various locations within, on, or about the heart and electrically coupled to the heart. In certain configurations, multiple electrodes may provide multiple sensing and/or stimulation sites within a single heart chamber. Theelectrodes505 are coupled to switchmatrix circuitry525 that is used to selectively couple theelectrodes505 to thesense circuitry510 and thetherapy pulse generators530,535.
• Cardiac electrical activity may be sensed from the electrode sites of the patient's heart are sensed viaelectrodes505 andsense circuitry510. Other types of signals, e.g., heart sound signals, pressure signals, respiration signals, and/or other types of signals may be sensed usingvarious sensors506 in conjunction withsense circuitry510. Sensed signals acquired using thesensors505,506 andsense circuitry510 are processed bysignal processing circuitry515 to enhance cardiac signal features used for cardiac event detection. The type of processing applied via thesignal processing circuitry515 may be based on the type of cardiac event being identified. For example, in one situation, the morphology of the cardiac signal component of the sensed signal may be altered to enhance a first set of features to facilitate detection of the cardiac pacing responses. In a second situation, the morphology of the cardiac signal component of the sensed signal may be altered to enhance a second set of features to facilitate detection of tachyarrhythmia or to identify a type of tachyarrhythmia.
• Altering the morphology of the cardiac signal component of the sensed signal may be accomplished by filtering the signal to remove certain frequency components. For example, high pass, low pass or band pass filtering may be applied. Additionally or alternatively, the cardiac signal morphology may be altered by changing data acquisition parameters such as A/D converter resolution or sample frequency. The signal acquired via theelectrodes505 orsensors506 andsense circuitry510 may also be filtered to remove or reduce the noise component from the sensed signal.
• The cardiac signal that has been modified to enhance feature detection is analyzed bycardiac event processor560. During this analysis, cardiac signal features present in the processed signal are used to classify various cardiac events. In one scenario, cardiac signal features present in a processed electrogram (EGM) signal are used by thecardiac event processor560 to classify the cardiac response to pacing. For example, the presence or absence of cardiac signal peaks within cardiac response detection windows may be used to determine whether or not the pacing pulse produced capture or some other pacing response such as noncapture, fusion or pseudofusion, or noncapture with intrinsic activation.
• In another implementation, the cardiac signal modified by thesignal processing circuitry515 may be used to classify cardiac tachyarrhythmia episode types. Features extracted from the processed signal of cardiac episode beats may be compared to template features that are representative of various types of tachyarrhythmia. The amplitude, timing, and/or other characteristics of the processed signal features of episode beats are compared to corresponding template features. If the episode beat features and template features are sufficiently similar, then thecardiac event processor560 identifies the cardiac episode as the type of tachyarrhythmia represented by the template.
• In some implementations, the parameters of thesignal processing circuitry515 may be selected based on population data. For example, a fixed filter configuration may be chosen to process the signal, where the fixed filter provides a desired morphology of signal features across the general patient population. In one capture application, a filter with a high pass corner frequency of about 3 Hz and a low pass corner frequency of about 100 Hz produces negative and positive signal peaks that occur with a desired amplitude and timing. Determination of the cardiac pacing response may involve sensing for the presence of positive and/or negative peaks within detection windows. For most patients, these signal processing parameters produce cardiac signals that provide accurate cardiac pacing response determination based on the location and timing of the cardiac signal peaks.
• As illustrated inFIG. 5B, in certain embodiments, thesignal processing circuitry515 may be implemented as a bank of signal processing units581-584, where each signal processing unit581-584 has its own set of parameters. For example, each signal processing unit581-584 may use parameters selected to enhance feature detection for a particular type of cardiac event. The cardiac signal from sense circuitry510 (FIG. 5A) may be processed by each of the signal processing units581-584. The outputs of the signal processing units581-584 are analyzed by thecardiac event processor560 to detect the presence of the cardiac events.
• For example, as illustrated inFIG. 5B, a firstsignal processing unit581 may be configured to enhance feature detection for normal sinus rhythm (NSR). A secondsignal processing unit582 may be configured to enhance feature detection for supraventricular tachyarrhythmia (SVT). A thirdsignal processing unit583 may be configured to enhance feature detection for a first type of ventricular tachyarrhythmia (VTA). A fourthsignal processing unit584 may be configured to enhance feature detection for a second type of ventricular tachyarrhythmia (VTB). The outputs of the signal processing units582-584 are applied to thecardiac event processor560 which may include circuitry for discrimination ofNSR591,SVT592,VTA593, andVTB594. In one implementation, thecardiac event processor560 may include circuitry591-594 to compare the signals processed by the signal processing units581-584 to templates representing NSR, SVT, VTA, and VTB. Identification of the cardiac events may be based on the degree of correlation between the processed cardiac signals and the templates.
• Due to disease state, patient pathophysiology, sensor characteristics and/or lead implant characteristics, the uniform parameter settings described above may result in sub-optimal feature detection for certain patients. In some embodiments, the signal processing parameter settings may be programmable via an external device programmer or other remote device management system. For example, in one scenario, a physician or other health care provider may select the parameters for signal processing and observe changes in the cardiac signal processed using the selected parameters. Through observation of the cardiac signal, the physician may determine if the selected parameters produce a desired improvement in cardiac signal morphology to achieve enhanced feature detection. In another scenario, the physician may determine if the modified parameters produce a desired effect, e.g., more accurate capture detection and/or identification of tachyarrhythmia episodes.
• In some embodiments, selection of the individualized parameter settings may be performed automatically by the device. For example, the device implement a process wherein the parameter settings are automatically incrementally changed and the signal produced using each of the parameter setting is analyzed. This process may continue until one or more undesirable features are reduced in the cardiac signal until desirable features are accentuated, or until optimal parameter settings are determined. For example, in one implementation, the device may identify an undesirable feature in the signal and then automatically step through the available signal processing parameter settings to determine whether, or the degree to which, each setting removes or sufficiently reduces the undesirable feature. If it is not possible to eliminate the undesirable signal feature from the cardiac signal, an optimal parameter setting which provides the greatest reduction of the undesirable feature is selected. In other implementations, desirable features may be accentuated through the use of automatic modification of the signal processing parameters. For example, the device may step through available signal processing parameter settings to determine an optimal parameter setting that produces a sharp signal peak.
• Methods to reduce undesirable features, accentuate desirable features, and/or to determine optimal parameter settings for signal processing may be implemented in firmware, software or a combination of both firmware and software. The automated parameter selection process may be performed on demand, periodically, and/or when degradation of cardiac event identification is detected. For example, if capture detection cannot be achieved or is intermittent, the signal processing parameter settings used for capture detection may be tested and readjusted.
• The flow graph ofFIG. 6 illustrates a method for automatically determining signal processing parameters for feature enhancement in accordance with embodiments of the invention. The parameter settings of the signal processing circuitry are initialized610. The cardiac signal is sensed and processed620 using the initial parameter settings. The processed cardiac signal is analyzed630 with respect to the presence of undesirable features that may interfere with cardiac event identification. If there is another signal processing parameter setting to be tested640, the parameters of the signal processing circuitry are modified660 to the new setting. The cardiac signal is sensed and processed620 with the new parameter setting. The processed signal is again analyzed630 with respect to the presence of undesirable features. This process continues until all parameter settings have been tested. The optimal parameter setting for feature enhancement is selected650 after all parameter settings have been tested.
• The presence of undesirable features in the cardiac signal may be determined using various techniques. Several exemplary techniques are described herein, but any technique that detects the presence of an undesirable feature or analyzes the comparative strength of desirable and undesirable features may be used.
• A technique using gradient analysis for determining the presence of undesirable features is described in the context of a cardiac signal used for capture detection.FIG. 7A illustrates acardiac signal710 having a morphology exhibiting doublepositive peaks720,730. As previously discussed, capture detection may rely on a signal peak occurring within a detection window having predetermined amplitude and timing ranges. The presence of a double peak may lead to uncertainties in capture detection, for example, when one peak falls within the window and one peak falls outside the window. Therefore it is desirable to eliminate the double peak by appropriate signal processing.
• In accordance with one embodiment, the presence of a multiple peaks in a cardiac signal may be detected using a gradient analysis. In this process, thegradient740 of thecardiac signal710 is determined. Thecardiac signal gradient740 is graphically illustrated inFIGS. 7A and 7B. Peaks in thecardiac signal710 are indicated in the gradient by zero crossings. Themain peak720 of thecardiac signal710 is indicated in thegradient740 by zerocrossing742. The additional peak730 in the vicinity of themain peak720 is indicated in thegradient740 by zerocrossing741.
• If one or more additional zero crossings are detected in thecardiac signal gradient740 at locations close to the location of the main peak zero crossing, as illustrated inFIGS. 7A and 7B, the signal processing parameters for this particular cardiac signal morphology are non-optimal. The signal processing parameters may be changed to eliminate or reduce the presence of the secondary peaks.
• FIGS. 7C and 7D illustrate thecardiac signal750 after it has been processed to reduce the double peak. The processedcardiac signal750 exhibits asingle peak760 that is indicated in thecardiac signal gradient770 by zerocrossing771.
• Another method for detecting undesirable features in the cardiac signal such as the multiple peaks illustrated inFIG. 7A involves the use of a curvature-based analysis. Through curvature analysis significant points of the cardiac signal are determined based on the curvature of the cardiac signal, along with peak detection. If additional significant points indicating secondary peaks having amplitude and location in the vicinity of the main peak, then the signal processing parameter settings may be modified to eliminate the secondary peaks. The use of curvature analysis to detect cardiac signal features including peaks and/or inflection points, for example, is further described in commonly owned U.S. Pat. No. 6,950,702 which is incorporated herein by reference.
• Modification of cardiac signal processing parameters to enhance feature detection in accordance with embodiments of the invention may be applied to improve discrimination of tachyarrhythmia events. In morphology based arrhythmia detection, feature points of cardiac signal are extracted, and the correlation with features from a template is calculated. If the correlation coefficient is higher than a predetermined threshold, the beat is classified as the type of beat represented by the template.
• In one implementation, the detected beat signal is compared to a template representative of supraventricular rhythm (SVR). If the beat signal is correlated to the SVR template, the beat is classified as SVR and therapy is withheld. If the beat signal is not correlated to the SVR template, the beat is classified as a ventricular tachyarrhythmia beat and therapy may be delivered. According to this approach, feature points1-8 are extracted from thecardiac signal810, as illustrated inFIG. 8. The use of eight feature points is based on the assumption that only 2 inflection points (feature2 andfeature4 inFIG. 8) exist in aregion820 around thepeak feature3 for a SVR beat. If there are additional inflection points inregion820, features2,3 and4 may not be identified correctly during template formation due to variation in determining these feature points. The rhythm analysis is further based on the assumption that there is no significant variation in curvature inregion830. If this is not the case, the local changes in morphology may cause variation in determiningfeatures2,3,4,7 and8 during beat to beat correlation as well as during template formation. The feature point variation caused by local changes in morphology may impact the ability to accurately classify the type of cardiac rhythm.
• In accordance with the approaches of the present invention, signal processing parameters may be determined that eliminate the undesirable morphology or curvature that occur inregions820 and830. The above-described gradient or curvature based methods may be used to detect the undesirable inflection points. For example, inflection points may be detected using the gradient approach as points where the gradient is zero.
• FIG. 9 illustrates the windows drawn around the QRS complex feature point locations (features2,3,4,7 and8) which serve to facilitate detection of significant changes in curvature and/or the presence of multiple inflection points within the window area. The signal processing parameters may then be selected to reduce or eliminate the undesirable features in the window area. The signal processing parameters may be optimized to produce the best correlation between beats from stored episodes and a template beat. If the signal processing parameters are significantly different after the adjustment has been made, a new template update may be initiated and the resulting template compared to the previous template stored in memory. If both templates are morphologically different, the new template may be stored in memory, replacing the old template.
• There are also other approaches to detect undesirable features or enhance desirable features in the cardiac signal component, such as using energy or power of the signal. For example, for the capture detection example, it may be possible to integrate the positive and negative deflections and choose the signal processing parameters which produce the smallest integral as the optimal parameter.
• Determination of the signal processing parameters may be performed based on patient population data or may be individualized for a particular patient. As previously discussed, the signal processing parameters may be automatically or manually determined. In one approach, the signal processing parameters may be determined during template creation and/or update. In another approach, the signal processing parameters may be determined based on the patient's stored episode data. In another approach, the signal processing parameters may be selected by a physician and uploaded to the device. In yet another approach, the physician could run a pacing or exercise protocol to simulate tachyarrhythmia with the device adjusting the signal processing parameters during the simulated tachyarrhythmia to enhance tachyarrhythmia detection.
• Various modifications and additions can be made to the embodiments discussed hereinabove without departing from the scope of the present invention. Accordingly, the scope of the present invention should not be limited by the particular embodiments described above, but should be defined only by the claims set forth below and equivalents thereof.

Claims (26)

1. A method for classifying a cardiac event, comprising:

sensing a signal comprising a cardiac signal component and a noise signal component;

processing the sensed signal to preferentially alter morphology of the cardiac signal component of the sensed signal, the altered morphology of the cardiac signal component enhancing detection of one or more features of the cardiac signal component;

detecting the one or more features; and

classifying the cardiac event using the one or more detected features.

2. The method ofclaim 1, wherein processing the sensed signal comprises processing the sensed signal using adaptable signal processing parameters.

3. The method ofclaim 1, further comprising automatically determining signal processing parameters used to alter morphology of the cardiac signal component of the sensed signal.

4. The method ofclaim 1, further comprising determining signal processing parameters that accentuate one or more desirable features of the cardiac signal component or mitigate one or more undesirable features of the cardiac signal component.

5. The method ofclaim 4, wherein determining the signal processing parameters comprises determining the signal processing parameters that enhance a correlation coefficient between beats from cardiac episodes and a template.

6. The method ofclaim 4, wherein determining the signal processing parameters comprises determining the signal processing parameters based on an energy or power of the cardiac signal component.

7. The method ofclaim 1, further comprising:

detecting one or more undesirable features of the cardiac signal component of the sensed signal; and

determining signal processing parameters that mitigate the one or more undesirable features of the cardiac signal component.

8. The method ofclaim 7, wherein detecting the one or more undesirable features of the cardiac signal component comprises detecting the one or more undesirable features based on a gradient analysis.

9. The method ofclaim 7, wherein detecting the one or more undesirable features of the cardiac signal component comprises detecting the one or more undesirable features based on a curvature analysis.

10. The method ofclaim 1, wherein classifying the cardiac event comprises classifying a cardiac pacing response.

11. The method ofclaim 1, wherein classifying the cardiac event comprises classifying a tachyarrhythmia event.

12. A medical system, comprising:

sensing circuitry configured to sense a signal comprising a cardiac signal component and a noise component;

signal processing circuitry configured to process the sensed signal to alter morphology of the cardiac signal component of the sensed signal, the altered cardiac signal component enhancing detection of one or more features of the cardiac signal component;

a feature detector configured to detect the one or more features of the cardiac signal component; and

a cardiac event processor configured to classify a cardiac event using the one or more detected features.

13. The system ofclaim 12, wherein the signal processing circuit comprises a filter configured to alter the cardiac signal component.

14. The system ofclaim 13, wherein the filter comprises a high pass corner frequency of about 3 Hz.

15. The system ofclaim 13, wherein the filter comprises a low pass corner frequency of about 100 Hz.

16. The system ofclaim 12, further comprising a noise filter configured to filter the sensed signal to reduce the noise component of the sensed signal.

17. The system ofclaim 12, wherein parameters of the signal processing circuitry are derived from population based data.

18. The system ofclaim 12, wherein parameters of the signal processing circuitry are determined based on individual characteristics of a patient.

19. The system ofclaim 12, wherein parameters of the signal processing circuitry are automatically adaptable.

20. The system ofclaim 12, wherein the cardiac event processor comprises circuitry configured to determine a cardiac pacing response.

21. The system ofclaim 12, wherein the cardiac event processor comprises circuitry configured to detect tachyarrhythmia.

22. The system ofclaim 12, wherein the signal processing circuitry comprises a bank of signal processing units.

23. A medical system, comprising:

sensing circuitry configured to sense a signal comprising a cardiac signal component and a noise component;

means for processing the sensed signal to alter morphology of the cardiac signal component of the sensed signal, the altered morphology of the cardiac signal component enhancing detection of one or more features of the cardiac signal component;

a feature detector configured to detect the one or more features; and

a cardiac event processor configured to classify the cardiac event using the one or more detected features.

24. The system ofclaim 23, further comprising means for automatically determining parameters for processing the sensed signal.

25. The system ofclaim 23, further comprising means for altering at least one undesirable feature of the cardiac signal component.

26. The system ofclaim 23, further comprising means for enhancing at least one feature of the cardiac signal component.

Images